Follow these steps:
1. Identify the reasoning
Target: the reasoning in the current context.
2. Break reasoning into atomic steps
Decompose into the smallest logical moves — one inference each. Make implicit steps explicit and number them.
3. Examine each link for validity
For each step: what is claimed, what it rests on, whether the inference is valid, and what unstated assumptions make it work.
4. Check the inference type
- Deduction: conclusion necessarily follows from premises
- Induction: generalizing from specific examples
- Abduction: inferring best explanation from observations
- Analogy: applying patterns from comparable situations
Verify the type is appropriate and correctly applied.
5. Identify gaps and leaps
Missing steps, unjustified jumps, connections assumed but unproven, circular dependencies.
6. Verify logical connectors
Confirm AND, OR, IF-THEN, NOT are used correctly. Map which conclusions depend on which premises to see how failures propagate and whether alternative paths exist.
7. Assess overall chain strength
Does the conclusion follow validly? Where is the reasoning strongest and weakest? What would strengthen weak links?